# Oral argument — Laboratory Corp. of America v. Davis > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A98591 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** April 29, 2025 ## Text We will hear argument next in Case 24-304, Laboratory Corporation of America v. Davis. Mr. Francisco? Mr. Chief Justice, and may it please the Court, two basic principles resolve this case. First, a class action is just a tool for aggregating claims. So if an individual plaintiff doesn't have Article III standing to bring his own claim, he can't bring it as part of a class either. That's why Lareau held that an intervener has to show Article III standing in order to intervene, and as Justice Scalia said in Shady Grove, class actions are just another species of joinder. Second, Rule 23b-3's predominance requirement leads to the same result. If a class is defined to include plaintiffs without Article III standing, and as a result you need thousands of mini-trials to separate the wheat from the chaff, the Article III issue necessarily swamps any common ones. This case is a perfect example. Plaintiffs who don't want to use kiosks don't have standing to challenge how kiosks work, any more than a vegan has standing to challenge how a restaurant defines a medium-rare steak. As a result, the Court needs to assess whether each of the 8,000 to 112,000 class members actually want to use kiosks, and that will necessarily swamp any common issues as the D.C. and First Circuits correctly held in the Rail Freight and Ascol cases. Plaintiffs' only response is to say that courts should assess the merits first and jurisdictions second. But that makes no sense. What if they lose on the merits? Either the unnamed class members are bound by a judgment regardless of whether the Court had Article III jurisdiction over it, or the Court has to determine if it had jurisdiction over each plaintiff in the first place. And that's why courts have to address the jurisdiction before the merits, just like in every other case. Plaintiffs' rule, in contrast, assumes either they win or coerce a settlement, but there's no basis for that heads-I-win, tails-you-lose approach to Article III. I welcome your questions. In this case, there have been a number of orders, and it seems as though the one that we have before us is the May order, which is inoperative. Would you spend a minute on why we have jurisdiction over the May order when there have been subsequent orders? Sure, Your Honor, for a variety of reasons. And I think the simplest way to understand it is that the Ninth Circuit resolved the question presented in this case, affirmed the certification of a class against us. It just got it wrong. You have jurisdiction to review that judgment. Now, I can unpack that a little bit more. Everybody here agrees that the May order is before the Court. The August order didn't change the May order in any material way. That's what the district court explicitly found. That's what plaintiffs argued to the district court below. That's why we couldn't have actually appealed the August order even if we wanted to. It would have been barred by 23F's 14-day statute of limitations. And I think, more importantly, that is precisely why the Ninth Circuit resolved the question presented in the context of the May order. It issued a judgment. That judgment went against us. It had reasoning. We think that reasoning is wrong. This court has the jurisdiction to address that judgment, reverse it, and send it back to the Ninth Circuit. Counsel, you're skipping a lot of steps in there. The May order is before us, but it's inoperative. The August order superseded it and replaced it. That's what the Ninth Circuit said when it reached the May order. It said, you didn't move to amend your notice of appeal. We're basing this on the May order. We're not basing it on the August order because you didn't move to amend. Isn't us looking at the May order, that's not the operative language right now. Isn't it an advisory opinion? Not at all, Your Honor, for a couple of different reasons. Well, you started by saying they're identical. I don't see them as identical. I see the two having very distinct differences. The first one is someone who's been denied the benefit of the existence of the kiosk. The second one is someone who, in my judgment, wanted to use the kiosk and couldn't. Now, you say the district court said they were identical, but the Ninth Circuit didn't say that. Well, Your Honor, a couple of responses. First, the district court explicitly said twice. I just said we're not looking at what the district court said. We have to look at what the Ninth Circuit said, and it said the August order is not before it. That's not true, Your Honor. What the Ninth Circuit said was that the August order isn't before it for purposes of an argument that we made specifically with respect to the August order on failsafe classes. The Ninth Circuit actually addressed the question presented under the May order, and it did so precisely because the district court repeatedly said that the August order did not materially change the May order. I think that's quite clearly what it did. I think another way to understand it, though, is that what the district court technically did was that it amended the May order. That's what the June order and the August order say. We're amending the May order. And the Ninth Circuit said you should have amended your notice of appeal. Only with respect to the failsafe class issue. Counsel, I still don't see how this is not an advisory opinion. The August order is not before us. Whether the Ninth Circuit was right or not in not granting you a right to appeal when you didn't ask for it, it's still not before us. I just think I fundamentally disagree with that, Your Honor. The Ninth Circuit actually addressed the question in the context of the May order, precisely because the August order didn't materially change it, as the district court said twice and as they agreed. But, Mr. Francisco, it does seem to have materially changed it, and it seems to have materially changed it along the lines of precisely what you're arguing about. In other words, one order has more uninjured class members in it than the other order. Because the first order basically said everybody who's injured by the ADA. And your whole objection to the second order is that it doesn't any longer say that. It just says, like, anybody who came into an office, regardless whether that person wanted to use the kiosk or not. So in the very way that you're saying the orders got problematically, you know, that the August order was problematically expansive, that kind of shows you the difference between the May order, which was confined to people who had ADA injuries, and the August order. So it's not just that we're kind of staring at the wrong order. We're staring at the wrong order, which is different in exactly the way that your arguments want to talk about. Well, I very much disagree with that, Your Honor, because everybody understood that the May order, the original order, actually included anybody who simply walked into a kiosk. But that's not what it says. It's got the fail-safe language, which you're not supposed to put in. You pointed that out, and the district court took it out. But you're saying you can't have a class unless everybody's injured, for Article III purposes. But by definition, the May order, improperly, had just such language. Your Honor, I think that the May order and the August order meant the same thing. If we start out with the May order. They don't say the same thing. Well, let's start out with the May order. One has a fail-safe, and the other doesn't. Sure, Your Honor, but when it comes to defining what the scope of the class is, here's what my friends told the district court with respect to the May order, the original order. That's nice that they said that, but that's not the class that the district court certified in May. Respectfully, Your Honor, it is. Because the reason why they said it included everybody that just walked into a kiosk is because their understanding of that definition, which the district court effectively adopted, was that anybody who was a patient at the facility and who happened to be blind was within the class. That's why the district court specifically said, I think a couple of times, it would be very easy to figure out who is in this class. We just should overlook the fail-safe language in the May order, then? Is that what you're asking? No, Your Honor. What I'm saying is the fail-safe issue is no longer before the court. That's the issue the Ninth Circuit declined to resolve. The issue that is before the court is the issue that the Ninth Circuit did resolve. What the Ninth Circuit said is that it doesn't matter on that issue whether the class includes people who haven't been injured, because under Ninth Circuit precedent, all you need is a named plaintiff with an injury, and it doesn't matter if there are other people who aren't injured. I understand how that principle could apply elsewhere, but why don't you give us the framework for understanding this? Because, just as Gorsuch is right, the language in the August order is different from that of the May order. You know, there's a bigger difference there than there is between May and June. So how are we supposed to figure it out? Are we supposed to look at it and say this language is different? Are we looking at it and saying, well, the principle that the Ninth Circuit addresses and that footnote is the same? I mean, how do you decide which is the operative order? So I think you can come at it in a couple of different ways. I think the simplest way is to say that the Ninth Circuit issued a judgment that certified a class, it adopted a legal rule that said it doesn't matter if there are any injured people in that class. You can correct that error to the extent you think it was wrong and send it back for the Ninth Circuit to figure it out after that. I also think you can get into it in a much more granular fashion. Here, the district court, which usually gets discretion over the interpretation of the scope of its own orders, the district court said that the August order and the May order meant the exact same thing when it comes to the scope of the... So I think the district court... I'm sorry if I interrupted you. Yeah, I was just trying to finish my answer to Justice's question. If the district court is right that the August order and the May order actually meant the same thing, and if the district court is right that all you had to do to establish the size of the class for either order was to simply ask, are these people patients of LabCorp and are they blind, which is what the district court said, then at that granular level, it's quite clear that as the plaintiffs repeatedly argued, the district court actually has adopted their view that anybody who had simply walked into the kiosk was injured. That was also the position that we took in our alternative argument. I guess the reason I keep on interrupting you is because you keep on saying what the district court said. And I think it's at least... I think the district court said many things. I mean, I'll grant you that the district court was a little bit unclear, but the district court also took the opposite position. I mean, this is the way the district court characterized the May class is legally blind class members who attempted to or were discouraged from using LabCorp's kiosks, and that's a quote. And what the district court was clearly saying there was people who wanted to use the kiosk, who went up to the kiosk, who couldn't use the kiosk. And again, that's the exact distinction that you are making is between those people and people who walked into the facilities but never wanted to use the kiosk. So I think that the district court understood its own order, at least sometimes, as going to a narrower set of people. Well, Your Honor, I think that the district court and the Ninth Circuit's opinions weren't models of clarity, but one thing was crystal clear in the Ninth Circuit and the district court. It certified a class based on a rule that it doesn't matter if there are uninjured people in the class. That's what the district court said. That's what the Ninth Circuit specifically said in affirming the district court's ruling. At a bare minimum, that Ninth Circuit judgment is before you and the rule that the Ninth Circuit adopted. That's, I think, why you granted certiorari in this case. But it's not an operative judgment. You're asking us to opine on the May 9th order that's been superseded by another one. And can I just point out, Mr. Francisco, that I guess I'm trying to understand your responsibility by you. I mean, your client's responsibility for the confusion that we are in right now. This is sort of what Justice Kagan is pointing to. I understood that your client argued that the May class definition was too narrow. You said it was improper because it was a fail-safe class. And as a result, the district court revisited it. And today you're saying it's improper or problematic because it's too broad. You know, it's too broad because it now includes uninjured people, whereas before it contained only injured people. And so I'm wondering if some element of estoppel isn't working here in the sense that you've taken opposite positions about what the problem is with respect to this class. Not even close, Your Honor. I think we addressed this fully in the letter that we just filed. We made alternative arguments in the Ninth Circuit. We argued it was a fail-safe, and then we alternatively argued this is Section 3A header. District Court manifestly erred in failing to consider evidence that many Rule 23b3 damages class members would lack standing to proceed. We then explained that at length in the brief as we quote in our letter. Lack standing because the May class, which was only injured people, was what? No, because the May class, this is what we said with respect to the May class, page 16 of our 23F brief. There is no evidence that the certified class contains a majority of persons or even a substantial number of persons with Article 3 standing. On the other hand, there is undisputed record evidence that around 25% of all LabCorp visitors choose to check in at the front desk and thus could not have suffered the injury required. Essentially the same argument that we're making here was an alternative argument. We lay this out very carefully in the letter that we just filed. Their assertion that we only raised issues with respect to the named plaintiff is wrong. We specifically raised the argument that we're presenting here. I think more importantly, that is the argument that the district court rejected. That is the argument that the Ninth Circuit rejected. And that is the judgment that is before you today. I don't think it matters whether it's the May order or the August order. All right, so why don't you tell us why is this wrong? Why are they wrong about it? I mean, just going to your Article 3 standing point, I appreciate the assertion that class actions are a species of joinder, but the absent class members are not participating. That's why they're absent. They're not parties in the traditional sense. So it seems odd to me that in a class situation where most of the time you don't even know who these people are, that's why you have the class mechanism operating, you would say that there has to be some showing with respect to individual injury at the threshold in this way. Sure, and there are a few responses, Your Honor. I think the first one is that their claims are before the court. That's why if the named plaintiff drops out because his claim becomes moot, the class action continues to proceed on the back of the live claims that are brought into a class action. But those claims are supported by the standing of the plaintiff who is named. Not if the named plaintiff's case becomes moot. No, but I'm not talking about mootness. I'm just talking about standing principles. Ordinarily, even if you had the names of every person or you knew all of the participants, only one person needs to satisfy the standing criteria in order to invoke the jurisdiction of the court. And so it seems a little at least discordant to suggest that in a class action situation, we have to figure out whether all of the unnamed class people, class members, have standing. I think that's wrong, and I think it's directly contrary to Lareau. But what I would also say is that I think that they could actually solve the Article 3 aspect of this relatively easily. They could simply redefine this class so that it is limited to people who, say, at a minimum, want to use the kiosk. They might need more, but at a minimum, want to use the kiosk. But that would just walk them straight into the Rule 23b-3 issue, because the only way to figure out the answer to that question is by having tens of thousands of mini-trials. That's why the response really does boil down to let's just deal with jurisdiction at the end, and we can deal with merits at the beginning. Counsel, in every class action, what you're defining as a mini-trial is what happens with respect to getting damages, meaning in every class action, a legal principle of liability is found, and then a mechanism is used to establish people coming in with their proof of injury. So that doesn't defeat class action. So I think the question is, is there an identifiable way to identify who's going to be a member of this class? Is there a mechanism that doesn't overwhelm the common questions? So, as I said, you're not going to have a class action if your definition is, are you going to have a thousand mini-trials? Every claim of damage is a thousand mini-trials. Is the mechanism manageable? That's the predominance question, and that's what they have to prove, which is, in my mind, and I'm not sure the court below did this, is there an identifiable mechanism that can identify who's part of this class? Certainly. And the mechanism has to be one that protects the defendant's rights. Also, the other thing I would point to is that unlike damages which come at the end of the case the right is not to have uninjured class members paid, and we still have mechanisms that decide whether they've proven their entitlement to a specific amount. That might be the right of damages. That's not the right of threshold, the threshold question. That's how Tyson Foods looked at this. It upheld the class-wide damages award and rejected the argument that the class should not have been certified, notwithstanding the fact that it was undisputed that the class contained hundreds of uninjured individuals. The court recognized that the Article III question of whether uninjured class members may recover is one of great importance, but it didn't view it as fairly presented by the case because the damage award had not been dispersed, nor did the record indicate how it would be dispersed. In other words, Article III requires standing by one plaintiff to get the jurisdiction of the court. Rule 23 requires that the common issues, not all of them, just some of them, be sufficiently predominant, and then the court can break off whatever it needs to break off. And I think that is not a proper reading of Tyson Foods. In Tyson Foods, the issue didn't even rise until the case went to the jury, and it rejected part of the damage expert of the plaintiff. But if you think about what happens under a rule that says we do merits first and jurisdiction second, again, ask yourself, what if the court certifies an overly broad class than rules against that class on the merits? The ordinary rule is that that binds the entire class. Well, that's not necessarily true. I think it's clearly not true if the class includes people over whom the court didn't have jurisdiction in the first place. So it basically means you're going to have to adopt a rule that either binds the class over whom you didn't have jurisdiction, obviously wrong, or after the fact, you're going to have to go through each plaintiff and decide whether or not you had jurisdiction over them in the first place. Thank you, counsel. Is there wiggle room in your theory? I mean, let's say you have a class that's roughly 5,000 people. Is it no good if there may be 10 people who don't qualify or 20 people? What's the margin of error? I think it's less of a numerical question and more what Judge Katz has said in the Railway case and what Judge Kayada said in the Asikal case. If they define the class at the front end such that it doesn't distinguish between whether or not people were injured, and the only way you can separate out the injured and the uninjured, consistent with protecting the defendant's rights, is by conducting, you know, thousands of mini-trials, that's necessarily going to swamp any common issues because that is a threshold question that always has to be resolved before you reach the merits. I think this Court's decisions in Halliburton and Amgem actually provide very good illustrations of this. What you said in those two cases was that if the fraud on the market theory failed in a way that would have required individual reliance determinations, that necessarily would have precluded class certification because it would have swamped any common issues. Well, this is that in spades. Here we're talking about a threshold jurisdictional claim that is, in federal court, an element of every cause of action that is brought before the federal judiciary that has to be resolved before the court reaches the merits, not afterwards. And here, if there isn't an administrable way to separate the wheat from the chaff, consistent with protecting the defendant's rights, it's necessarily going to devolve into those mini-trials. Thank you, Counsel. Justice Thomas? You probably covered this at some point, but could you just spend a minute on why the May class definition includes so many uninjured? Sure. It's because the way the May class reads is it says anybody who, you know, it kind of parrots the language of the statute. But their understanding of the language of the statute that the district court adopted was that that includes anybody who walked into the facility. Again, this is what they said, my friends said in their brief to the Ninth Circuit, when we were all talking about the May definition. They said that standing is established for all legally blind Californians who visited one of the 280 facilities that featured a kiosk. They then explained that the way that you identify who's in that class is by simply looking at LabCorp records that show who visited it, who were their patients, and you match that up to other medical records that show who was blind. The district court adopted that, and we know that for two reasons. We know that, one, because the district court said that that's exactly how it would define who was in the class, and two, because it adopted the August definition that clearly does that and then told everybody that that August definition meant the exact same thing as the May definition. That's why, Your Honor, we actually could not have appealed the August order if we wanted to. The binding rule in the Ninth Circuit is that if a subsequent certification order makes immaterial changes to a prior certification order, the 14-day statute of limitations runs off of the prior order, not the subsequent one. The district court having squarely held that the August order was not materially different than the May order, it would have precluded us from appealing the August order and left us stuck with the May order. The reason none of this mattered is because everybody agreed that the two orders meant the exact same thing. To the extent there was any daylight, the district court went further and said the August order is simply amending the text of the May order. That's why, while the Ninth Circuit said I'm not going to address any issues that pertain only to the August order, and that was the fail-safe issue, it argued, it claimed, I think erroneously, but it claimed that our fail-safe argument pertained only to the August order, but the Ninth Circuit did address the very question that we've asked this court to resolve. It just got it wrong. It resolved it against us. That's the decision that's before you. You clearly have certiorari jurisdiction over that decision for the reasons that my friend in the previous case very well articulated when he was standing up here. At what point does having uninjured plaintiffs in the class present a problem for Rule 23 or for Article III? Sure. So I'll take them in order. I think for Article III, if the class on its face is defined such that it clearly sweeps in uninjured people, I think that you can't certify that class. You've got to tell them to redefine the class so that it's limited to people over whom there's jurisdiction. Again, I think that can be solved relatively easily in a case like this by simply redefining the class to include, you know, at a minimum, people who wanted to use the kiosk. The problem with that is it walks you straight into the Rule 23B3 issue, where the question is, when you've got an overly broad class definition, is there a way that you can separate out the injured and uninjured in an administrable way? I'll give you an example that illustrates it. Take the TransUnion case. In TransUnion, you had a couple of thousand out of 6,000 plaintiffs whose credit reports actually were disseminated to third parties. If they had come in, as they originally did and said, the class includes everybody for whom there was a violation, a class of 6,000 people, that would be an overbroad class, and I don't think you could certify it. But what you could say to them is, you need to redefine the class. Redefine the class to include only those individuals whose credit reports were disseminated to third parties. Then you'd have a class that's defined to include only people who suffered an Article III injury. It would satisfy the Article III issue. You'd then have to move to the 23B3 issue and ask, is there an easy way to separate the wheat from the chaff? In that case, there probably would have been. You probably could have just looked at TransUnion's records to separate the wheat from the chaff. But in lots of cases, there is no easy way. That was Judge Katsas' opinion in Railway, Judge Kayada's opinion in Asikal, and this case. Because in this case, the only way that you can determine whether at a minimum somebody even wanted to use one of those kiosks is by putting that person on the stand. It's after all quite obvious that there are many people in this world who don't like to use kiosks. I happen to be one of them. If LabCorp adopted a policy that said that 5'8 overweight Filipino-American men have to use the front desk and not the kiosk, I would say, hallelujah. That might violate somebody's rights. It doesn't violate my rights. And the only way to figure that out in this case is through doing the thousands of mini-trials. Thank you. Justice Alito, anything further? Yeah, a few questions. I'm troubled by the Ninth Circuit's rule that a new notice of appeal must be filed or a prior notice of appeal must be amended if the district court, while a class certification appeal is pending in the court of appeals, makes a material change in the class certification, but presumably not if the district court makes an immaterial change. And that's where we've spent most of the argument this morning on that issue. But whether someone has to file a notice of appeal is supposed to be clear, and that is such an unclear rule. I'm wondering whether that's the root of the problem that we've been discussing. But do you seem to accept the proposition that we should view this matter through the lens of this Ninth Circuit rule about what needs to be done? Well, Your Honor, I don't necessarily accept the premise, and I think that might be another way to go about it. But what I do think is that two things. One, the Ninth Circuit squarely addressed this question, and the reason it squarely addressed this question was because under the rules that the Ninth Circuit applies, this issue was squarely before it. The May order was before it. The Ninth Circuit and the district court repeatedly made clear the August order didn't materially change anything. That meant both of them presented the exact same issue. That, again, is why the Ninth Circuit said, I'm not going to address an issue that pertains only to the August order, and it didn't. It refused to address that. But it did say, I am going to address an issue that relates to the May order, because on that issue, there is no difference between the May order and the August order. They present the same thing. I think, though, that the way you cut through all of this is you recognize that this court is a court of review. It's reviewing the Ninth Circuit's decision. The Ninth Circuit entered a judgment against us in which it adopted a legal rule. That legal rule, in our view, is wrong, and you have jurisdiction under the certiorari statute to say whether or not the rule that the Ninth Circuit adopted. The rule for filing a notice of appeal, when a district court keeps changing its class certifications, could be you always have to file a new notice of appeal, or you always have to amend, all right. It's a pain, but that's what you have to do. Everybody would understand it, so everybody could comply. The rule could be, no, you never have to do that. Once there's a certification, the case is in the court of appeals. If the district court can't make up its mind or gets worried about the possibility it's going to be reversed on appeal, well, that's too bad. Whatever the latest certification order is before the court of appeals, unless the court of appeals chooses, as a matter of discretion, to dismiss the appeal. Either of those would be clear. But what we have in the Ninth Circuit is something that is utterly unclear, the drawing a distinction between a material change and an immaterial change. I think that is a fair general criticism. I do think that in the application in this case, it's quite easy, because here the rule is you can't appeal if there is an immaterial change. You're stuck with the original order. And here we have the district court saying twice, in the very orders themselves, this is not an immaterial change. It could not have been clearer in stating that, and therefore making clear to us that the only avenue was to challenge the May order. I also think that that makes sense in the context of this case, because my friends on the other side repeatedly told the court that the May order encompassed anyone who had simply walked into a kiosk, then when they sought to have the new August order implemented. We didn't ask for the August order. They asked for the August order. When they asked for the new August order, they again told the district court, this is just a housekeeping function. It's not going to change the scope of the claim. So everybody agreed they were the same. One question on the merits. This court has said that non-named parties are parties for some purposes, but not for other purposes. And I know you want us to hold that they must be parties for Article III purposes. But if you step back, can you offer any sort of general rule for determining when they must be regarded as parties and when they need not? I think my basic answer is I don't think that it really matters, because what you're adding are claims to the case. Whether you're adding parties or not, their claims are clearly being added to the case, and you're being asked to adjudicate those claims. And I think that's really what you were getting at in the LaRocle case, Your Honor, when you said that any time you add a new claim to the case, you have to have Article III jurisdiction over that claim. And it goes back to the exchange I was having earlier, where if the name plaintiff's case becomes moot, and that plaintiff drops out as a party, a class action continues to proceed if there are claims from absent class members over whom the court has Article III jurisdiction. That's the critical issue. Thank you. Anything further, Justice Sotomayor? Counsel, I want to follow up on your answer to Justice Alito. When does a party become a part of the litigation? I always thought it was at the time the class was certified, which is not at the time where the judgment is entered. I didn't think they became parties until the judgment is entered in a class action. I think their claims are clearly added the moment the class is certified. To claim the claim, yes, but not them as parties. So I'm not quite sure why they have to. They have to prove that they are injured or uninjured. So I'll give you two answers. The first is I think just as a technical, formal matter, when the claims are added at certification, you have to have jurisdiction over those claims. The practical answer is the one that I think I've already given, which is what if the class is certified, all the claims are added, and the court then rules against the class. The rule is the whole class is bound. Counsel, you said earlier, yes, but it's not bound until the class is certified, and between, until judgment is entered. The way class actions happen, they get amended constantly during the procedure. Sometimes some claims are dropped altogether. They're altered. The whole process is fluid as problems start arising, and it's not until the judgment is entered that you have to determine whether there's an administrable way to identify, I'm using your own words, to identify who's been injured or not. So you're saying instead we've got to do it immediately. They're saying it has to, the court below said it has to be done. What I'm saying is that you need to always address jurisdiction before you adjudicate the merits of the claim. Well, but we have jurisdiction. We have some people, the named plaintiffs, who wanted to use this kiosk. They are clearly a part of that class. Now the question becomes in naming that class, will there be people who are blind who didn't want to use it, and they have to show that there's an administrative way to identify the difference between those people and themselves. So, Your Honor, let's assume for the sake of argument that people who don't want to use kiosks don't have standing to challenge kiosks, and let's further assume that we've got a class here that includes both groups. Oh, but that's going to be the legal fight. No, I get that, which is why I'm just stipulating. And I'm asking why do it immediately at class certification stage. There is an argument that if I'm blind, the benefit that the statute gives me is like the standing question we faced with the ADA person who was calling hotels, and all she said is, I want to walk in there, I may want to use it. We didn't require her to use it. All she had to say was, I may want to use it. They're saying the same thing. If I'm blind and I walk in, I'm being denied the choice. I don't know if that will hold up for injury. Sure. But the question is, why are we facing that question at class certification? And that's the question I want to answer, because I think what your question really is doing is joining the issue, and the issue that needs to be resolved. Mr. Francisco, hear me out. I think the question only becomes pertinent when you're trying to give a damage award to anyone. And that's where I fundamentally disagree with you, because that's only true if they win. If they lose, if the class is certified and they lose, the entire class is bound by that adverse judgment. Well, it depends on what ground they lose. Let's say they just lose across the board on the merits on a motion to dismiss. That binds the entire class. So if you have certified a class that includes uninjured people and the class loses across the board, the general rule is the entire class is bound. But that can't be the case if you didn't have jurisdiction in the first place, which is why jurisdiction always, in every case, precedes the merits. It doesn't follow the merits. That's one of our principal objections to their position. Your position, Your Honor, as you just very well articulated, makes perfect sense when they win, but it makes no sense when they lose. But you're still saying it's a question of predominance and that that's what has to be addressed. If there's an administrable way to do it, then the class gets certified. If there's not, then you can't. So I'm saying two things. Under Article III, they've got to define the class properly at the front end. Under 23B3, I'm not sure I necessarily disagree with you, but you've got to have an administrable way to separate the wheat from the chaff before you address the merits, not after you address the merits. And if the only way to do that, consistent with protecting the defendant's rights, is thousands of mini-trials, that is necessarily going to defeat predominance, just like if the fraud-on-the-market theory fails in a way that requires individual issues of reliance. Again, the rule that we're asking for under 23B3 is the one that Judge Katz has adopted in the D.C. Circuit and Judge Kayada adopted in the First Circuit. Justice Kagan? I want to pick up there and also go back to the Chief Justice's question and make sure I understand what you're saying, which is that the Article III question you're saying in a case like this is not so hard to solve, that it's merely a matter of taking it from all the people who walked into the facilities to the blind people who walked into the facilities and wanted to use the kiosk. And that's the key move, to get you to a place where it's not the Article III question that's important, but rather the predominance. With one insignificant tweak for purposes of this, I don't know that we would concede... In fact, I know we wouldn't concede that merely wanting to is good enough. I think there has to be a further injury, but it doesn't really affect the force of your question. OK, wanting to tried to something like that. And was hindered in their ability to check in. Yeah, OK. Then we can go back to the procedural question again, which I don't want to do, but that does raise the procedural question, and the procedural question is like, isn't that exactly what the May order said? And the problem was really the August order, which was way wider than that, but the May order seems on its face, and at least in some of the district court's comments, to be exactly that. Except for when the district order specifically addressed it, explicitly told us there's not an iota worth of difference between the two orders when it comes to the size of the class. I totally concede that it does say that at some times, and then it says at other times, I'm just talking about people who tried to use the kiosk, so that's a little bit of a mystery. But the one time it specifically did address it, twice, it said the two orders are materially identical. Yeah, on its face they're so obviously not materially identical. But, OK, I actually was going someplace else. We'll leave that as a question. OK, but that's the full range of the Article I question you see, and everything else you're saying really is a predominance inquiry, and is a matter... And I think you said it's not how many people. It's just you have a mechanism for easily separating the wheat from the chaff, is that correct? While protecting the defendant's rights. Yeah, well, that is what protects the defendant's rights, isn't it? Is there something else that I'm missing? Well, you know, I mean, there have been suggestions that you could just do everything through affidavits and deprive the defendants their right of contesting those affidavits through the crucible of cross-examination. It's specifically what Judge Kayada addressed in the Asikal case, where he said, look, when it comes to issues of preference, you can't simply rely on affidavits, because unless the defendants stand up and say, we concede that all of the affidavits are true. Instead, a defendant, if it has a good faith basis to do so, has a right to contest the veracity of affidavits. Here, given how implausible it would be if 112,000 people came forward and said, we all prefer to use the kiosks, given the inherent implausibility of that, I think we would have a very strong basis to say, no, we want to test every one of those affidavits. We're going to spend anywhere between two and ten years addressing the threshold question of jurisdiction. Necessarily going to overwhelm any common issue. So you're saying you can't do it by, like, surveys or other mechanisms that address the class as a whole? Those only work if you could also introduce them in an individual case. That's what you held in Tyson's. But take an individual case like this one. If somebody came in and said, I like kiosks, I want to use the kiosks, you couldn't put forward a survey that says 50% of people like to use kiosks, therefore you like to use kiosks. That would be completely inadmissible in an individual case, so inadmissible in a class case. So what they have to do is identify something that they could do that's consistent with our right to cross-examine and contest that evidence. And in a case like this, there's simply no way to do that, short of putting these people on the stand and testing whether their assertions survive the crucible of cross-examination. Okay, that's helpful to me, and I hope that Mr. Gupta addresses the same question. I'm sure he will. Because what I hear you saying in your argument, again, aside from the procedural matter as to whether you have any rights to object to the May order, is really that the crux of the matter is something along the lines of there's no procedure that we can think of that is easy enough to address the predominance inquiry. Well, again, it's not just any procedure we can think of. It's a procedure that exists and also is protective of the defendant's rights. Again, this is an issue that I think Judge Katsas and Kayata both went through in quite a bit of detail. Yeah, I didn't mean to take that out. I meant to say exactly how much does the defendant need such that you can have a mechanism that actually works to separate people who are injured from people who are uninjured. And the critical issue I would add to that, Your Honor, is that that is a procedure that also always has to be capable of taking place before you reach the merits. You can never kick the jurisdictional question to after you resolve the merits. It's also always got to be resolved before you reach the merits. And this is only because you're worried about the case of you if the defendant loses. I think that just illustrates the nature of the problem. I think that just reflects the fundamental principles of Article III jurisdiction. When a class action is just an aggregation tool. So when you certify a class, you're adding a bunch of claims to the case. You're increasing the exposure to the defendant. You cannot, you don't have the power to adjudicate the merits of those claims either up or down unless you have jurisdiction over those claims in the first place. Yeah, I think this one, I think we're going to have to agree to disagree on this one because the court is not doing anything with respect to those claims until the court actually provides damages, otherwise exercises remedial powers with respect to those claims. And as long as the court figures this question out before the court actually does anything with respect to those claims, that seems to be good enough. Otherwise, they're just riding along. They're not affecting the litigation in any way. And I disagree with that for a couple of reasons. One is that as this court has said a couple of different times, class actions are claims aggregation tools. As Justice Scalia explained in Shady Grove, it's just another joinder device. And I think that the reason why I point to what happens if there's a loss is because it does illustrate that the claims are, it illustrates the basic principles that I'm trying to articulate in other ways because it shows that those claims actually do become part of the case at the moment of certification because they are capable of being resolved in an adverse way against the defendant. And I think that is why this court has always said that class actions are just joinder devices, no different from intervention. And LaRoe makes clear that if you're going to add that new claim to a case by way of intervention, you need to have jurisdiction over that new claim. I don't think there's any different of a rule that would apply in the class action context. Thank you. Thank you. Justice Gorsuch? Sorry to belabor this, but I am struggling to understand your argument. I believe in response to the Chief Justice, so you can correct me, that you acknowledge that a court can certify class action with some non-injured people in it. And that in response to Justice Kagan, you said basically that you have to do the predominance and manageability inquiry early on. That's what I got out of it. And determine whether you can separate the wheat from the chaff early on in order to ensure that you can weed out people who aren't injured. And if all that's true, and you can tell me where I'm wrong, boy, that sure sounds like Rule 23 to me. So, Your Honor, I think I wasn't clear if that's how you understood my articulation of the rule. There are sort of two steps. There's the Article 3 step and the 23b3 step. Yeah, and the Article 3 step... You need to define the class so that it's limited only... Only so you can never have an uninjured person in a class definition. At least not on the face of the definition. But I think that in a case like this, and in most cases, that can be solved pretty easily. You just define the class... I've heard all that before. I don't mean to force you to repeat it. So your position now is a class definition can never have one uninjured person in it. If I can imagine a definition that yields one uninjured person, I can't certify it? Well, if you can imagine a class definition that yields one uninjured person, you can redefine the class to eliminate that uninjured person. Well, maybe I can and maybe I can't, but I know that common issues predominate, and I know that I can sort out those things later, and I still can't certify it? I think what you do with the proper approach there would be to simply redefine the class to eliminate the one person. You can say, okay, the class... So I think the answer to the question is yes. You cannot certify that class at all. Yeah. But you can redefine it. And if that's true, is that protective of defendants' rights? I mean, we've been talking about judgments in class action litigation. My memory's a little hazy, but I remember doing a little bit of it back in the day. And these things never go to judgment. They're always settled. And often defendants like broad class definitions because it gives them peace. And the alternative, which I think your rule would invite, is mass tort claims in which you're litigating these. You talk about piecemeal. You're really going to be litigating at piecemeal. Thoughts? So I guess my first thought would be if you look at just as a practical matter are the positions that we're articulating pro-defendant or anti-defendant. I guess my first answer would be I don't think it really matters. But my second answer would be that to the extent it does, I'm pretty comfortable with my position from a pro-defense standpoint because if you look who's lined up in favor of our position, it's pretty much the entire defense bar. Then to take it on more directly, what I'd say is if you can properly define the class, the case can be easily settled. You just have to figure out who's in that class and settle it with respect to those people. I think that the problem when you can lard up a class, not just with one, not just with two uninjured members, but you can define a class in a way like this one that maybe includes as many as a majority of uninjured members out of the 112,000 people. Then you'd maybe have some predominance issues and some manageability issues. I take all that point, but that's what Rule 23 exists to sort out, and maybe it isn't certifiable for that reason. But that's Rule 23 inquiry, it seems to me. And that's the second part of our argument. Even if you completely disagree with me on my Article 3 question, on 23b.3, if there isn't a way to separate them out before you reach the merits, short of having all of the mini-trials, it's going to fail under 23b.3 too. Thank you. Justice Calabro? I thought one of the problems with an overly broad class being certified was that it would pressure defendants into settlements that are coercive and unfair. Isn't that one of the concerns you have? Yes, Your Honor, and that would have been fourth on my list had I had a chance to get to that. And the other point I'd like to make... Do you want to explain that, the real-world problem? Yes, in the real world, what drives settlement is the fact of certification and the size of the class that's certified, because those are the two numbers that really require defendants to roll the dice. Maybe I can defeat everything at the end of the day. If I don't, I'm looking at a massive number times whatever damages there are per person. So the flip side of the piece from a larger class is the disaster, from your perspective, of being pressured into a settlement with an overly broad class once it's certified. I think that's right. We don't want to be pressured into those settlements. And the other thing that I would add is it's not like you need class actions across the board in every manifestation in order to make sure that plaintiffs are protected and defendants are punished. You also have federal and state law enforcement authorities who are charged with enforcing the federal and state consumer protection and anti-discrimination laws. One of the problems with an overly aggressive use of class certification is that it interferes with that law enforcement discretion by deputizing literally thousands of plaintiffs' lawyers to act as private attorneys general. I think this case is a pretty good illustration of that. Here, about a year ago, the Department of Health and Human Services actually put forward a rule that suggested that what we are doing is what we should be doing, that is, providing a front-desk alternative to kiosks. Yet notwithstanding that rule, we're being subjected to a massive class action that goes after us for doing precisely what the rule appears to contemplate. And on the facts here, I think you've maybe covered this, but I just want to be clear. On the facts, general facts here, could they permissibly define a damages class consistent with Article III and 23b3, and if so, how? I think they could do it here with respect to Article III. On these facts, I don't think they could do it consistent with Rule 23b3. They could. Let's assume that anyone who wants to use a kiosk has standing. As I mentioned to Justice Kagan, we dispute that, we think more is needed, but for purposes of this, I'll assume that's enough. They could define the class as anybody who wanted to use a kiosk and visited a LabCorp facility and couldn't use the kiosk. That then walks you straight into Rule 23b3, and I don't see any way for them to show, to meet Rule 23b3, because you'd have to have literally 112,000 mini-trials to determine whether any particular unnamed member actually wants to use that kiosk, particularly given how many people in this country, I talked about myself, I imagine I'm not alone in this room, don't like using kiosks and will avoid them whenever they can. Thank you. Justice Barrett? I'm with you. I like to avoid kiosks too. Okay, so I want to figure out exactly what will be open to you on remand. I mean, there are different paths that we could take here. Let's imagine that we say, oh, this whole confusion about the order, we think that the August order is operative and that's not before us, so we send it back. Given the Ninth Circuit's rules and that you have a time, like you were pointing out under 23f, you had time and the Ninth Circuit says that you couldn't appeal the August order, right, because they said it was the same. You say the Ninth Circuit, and you're right, that the Ninth Circuit said you couldn't appeal the August order because it was the same. So have you lost it? If your friend on the other side is right and so it's not properly before us, can you still appeal that certification? Do you follow me? Am I being clear? I think so. Okay, so if it's the case that what we really have before us is the May order or the May order as amended in June, and if we said, no, no, no, no, no, really it was this August order, you know, Justice Kagan was pointing out, no, I think the language is materially different, let's say that that's the view that carries the day. What happens to you? Are you still able to make these arguments with respect to the August order? So if the May order was immaterially amended by the August order as the district order said and the Ninth Circuit found, no, we cannot appeal the August order. But what if Justice Kagan is right? You know, she said if you look at the August order, Justice Kagan's question to you was, and I know you disagree with this, so just assume this. Let's assume that we thought that the August order did materially alter the May order. What happens to you? All right, so the assumption is that we're going to override the district court's own interpretation of its own orders. I get it, I get it, and override their understanding of the orders. I'll accept that too. I would have to think through that more, Your Honor, and I would be uncomfortable making a definitive representation here given how far we are away from all of the different orders. If you did that, I would certainly probably do my best to come up with an argument that we could appeal that August order separately. I don't think that there's any reason for you to do any of that because I think the simplest route here is that you have a Ninth Circuit judgment before you. I understand that, but I think you can tell from some of the questions today that there's some question about that on the bench, so I'm just trying to figure out what happens if that doesn't carry the day. Justice Alito pointed out this is a weird, not clear rule from the Ninth Circuit, so I'm trying to figure out what the consequence to your client would be if some of those concerns carried the day. I understand it's not your position and there are other routes open. The reason why I think it's complicated is let's say you reversed the Ninth Circuit's procedural ruling and you said the August 8th order was the operative one. Under Rule 23F, we are way past the 14-day period to appeal the August order, so there would have to be then some kind of equitable tolling concept that gets built into it on top of that, and as I said, if you were to do that, and I would strongly urge you not to, I would be vigorously arguing for anything I could think of to allow us to appeal that August order well past the 14-day period of limitations under 28F, and I would do my best to succeed. I just can't represent to you what I think the answer is. No, I agree, and that's why I brought up the 23F time. So it seems to me that maybe, and tell me if you think this is a description of your dilemma, if we understood the August order to materially, despite the descriptions in the lower courts, if we understood it as Justice Kagan was hypothesizing it, there was a material difference. We said, no, no, no, we've got to look at the orders in front of us. That's wrong. The whole reason you would be in this position is because of the weird rule that Justice Alito was pointing out, this not clear rule. We would be sending it back, and then it would be kind of, there would be a risk of too bad for you because the 23F timeline has run. So another way to look at this would be for you to say, okay, there might be some procedural quirks. Maybe they flow from the Ninth Circuit's odd way of deciding what orders are appealable, what orders are operative, but that here you should just decide the question presented on the facts as they've come up to you because you do have a judgment in front of you from the Ninth Circuit. A hundred percent. Okay. How much of this, I mean, I take it, I just want to clarify, in your comments to Justice Kavanaugh, we didn't take the case to decide whether, in fact, under the class as certified by the district court, there would be standing. Justice Sotomayor, I think I heard her say that we had decided that the woman who called the hotels had standing, even though she didn't walk in. We actually didn't in Acheson reach that question, and we didn't take this case to decide that here. But that is still open to you to argue on remand. Yes, Your Honor, and the only thing I'd add to that is I think that the facts here provide a good, nice way to illustrate the application of the rule because I think the facts are relatively clean, notwithstanding the procedural issues that we're discussing. But, yes, it would still be open to us on remand because the rule that the Ninth Circuit and the district court adopted was that it just didn't matter. It didn't matter whether the class included uninjured people because under the Ninth Circuit rule, you can certify a class as long as the named plaintiff has standing, and it doesn't matter if there are lots of other people who don't. Got you. Yes, I agree, and I think that's why we took the case to decide that issue and not just kind of carving out that other issue and saying that you are not accepting that this class definition that everyone in this class could satisfy Article III even if you collected 100,000 affidavits that said we walked into the lab core, didn't matter if we wanted to use the kiosk or not, but we couldn't have used it if we wanted to because we were blind. We would not accept that as a valid class. Yes. Yes. Okay. Thank you. Justice Jackson? So it seems to me that the merits of your argument actually rests on two premises that I am struggling with, so maybe you can help me. All right. I hear you saying at bottom that it violates Article III to include uninjured people in the class definition and that it violates Rule 23 if there are lots of uninjured people in the class definition, and so if I can just ask you questions about those two different basic propositions that I think is really what is underpinning your arguments here. The problem that I keep coming back to with your Article III point that it violates the Constitution to include uninjured people in the class and so you would therefore need to redefine it to be only injured people is our standard rules and principles with respect to standing and when the jurisdiction of the court can be invoked. My understanding is that you only need one plaintiff, one plaintiff who establishes standing even if there are others there who are making the same claim. I appreciate that our law says that they're making different claims by nature. I mean, obviously they're different because there are different people there, but what we say is if there's a claim that is being made and the claim is you violated the law in this way and we have five people who are saying that and there are named plaintiffs in this action, only one of them has to establish injury for standing purposes. If that's true, I don't understand your Article III argument. So I would push back on whether or not that is true. This court has never applied the one plaintiff rule to a damages case, which I think because by definition in a damages case, every plaintiff is seeking his own form of damages. And we've done that at the threshold? I mean, my understanding is that... That's the rule. No, my understanding is that, yes, at the end of the day, each person has to have been injured in order to be entitled to damages, but for the invocation of the power of the court, which is what Article III standing is about, we don't go into the harm to each person in order to take up the claim that is being made. I respectfully disagree with that, and I think it's squarely foreclosed by LaRoe. What LaRoe specifically said was that at the point of intervention, you don't allow the intervener to add his new claim to the case unless he can show an... But I'm not talking about intervention. I'm talking about original action. We have five plaintiffs. They are making a claim. They have one count in their complaint, and I understood that many, many times we just say, one person, show us your harm. Never in a damages case have you ever done that. Only in injunctions, and even there, only where plaintiffs receive the same injunction or declaration. I understand. That's where I'm having the disconnect. All right, the second problem is with respect to the proposition that it violates Rule 23 if there are lots of uninjured people in the class, and I've got to tell you, I'm struggling with why it matters that there are uninjured people. I hear you say that the reason is because we have to have a bunch of mini-trials, and I just want to put to you a quick hypothetical, which to me demonstrates that that's not always the case, and so, therefore, that might be a problem with your argument. So suppose we have a Verizon customer who brings a class action against the company, arguing that Verizon charged her and all customers certain fees over a six-month time period that she says were unlawful, and this is a claim that does not have an element of harm in it. She's just saying these fees, unlawful, you weren't allowed to do it, and she seeks to certify a class of all Verizon customers during that six-month time frame. Now imagine that Verizon says that it only charged some of its customers during that six-month time frame the relevant fee. So in actuality, only some of the members of the defined class were injured, and Verizon says that over time, with some effort, it can generate a list of those customers. So at the class certification stage, everybody knows that we will eventually be able to figure out which customers were actually charged the fee, but we have a class that's defined of everybody during the six-month period. I guess I don't understand why it matters how many injured versus non-injured members there are in this class as defined. Yeah, I think the way that the resolution of that hypothetical would proceed was that the front end, if you know that you've got a class that includes both people who were charged the fee and people who were not charged the fee, you define the class to include only people who were charged the fee. That solves the Article III problem. Well, assuming there is an Article III problem. Right, assuming there is an Article III problem. I totally agree with that. Then you get to the second stage, and you do the Rule 23b-3 analysis, and you say, and it's not really, as I mentioned to the Chief Justice, it's not so much a numbers game. The question is, is it easy to figure out who paid the fees or not? And in your hypothetical, it might well be very easy to figure it out. So we do certify that class, or we don't? I think that it would turn on how easy it is to figure out definitively who's in the class. But why? What difference does it make when we're certifying this class to establish the liability? There are common issues with respect to that. And really, the only thing that figuring out who is harmed and not matters to is who gets damages at the end of the day. I don't understand why class certification would be held up or evaluated with respect to the numbers of people who were actually injured or not in the class. And with respect, Your Honor, I think it makes all the difference in the world from both a practical matter and a legal matter. From a practical matter, these bloated classes are what allow plaintiff's lawyers to extract massive settlements on weak claims. From a legal matter, what you are doing are adding claims to a case over whom the court doesn't have jurisdiction. Assuming your Article 3 question is correct. No, even assuming my Article 3 question is wrong, Your Honor, if you have a class that includes people who have not been injured, I'll assume that you don't think that that is an Article 3 problem. When it comes to Rule 23b-3, you still at some point have to figure out whether or not you have jurisdiction over those individual claims. And you cannot proceed to adjudicate the merits of those individual claims unless you first assure yourself that you have Article 3 jurisdiction over the individual claims. One final question. Assuming the claim is different from other claims, but setting that aside, with respect to the practicality of it, I appreciate Justice Kavanaugh's point that many of these settle and that it sort of tilts the scales in some way from the defendant's perspective, but don't defendants also have, in my case for example, all of the information that would be necessary for them to say, we know that only X number of people have injury. In other words, the defendants have the best lawyers, they have a gajillion dollars, they are being sued, and they have some responsibility and understanding of the claim and the population of people who were injured, right? And I think that's why, Your Honor, in your hypothetical, I said that it would be pretty easy to define the class that met our Article 3 rule. Anybody who paid the fee. And on the 23b-3 issue, in a case that really is just looking at the company's records to figure out who paid the fee, that might well survive the 23b-3 inquiry as well. It's essentially like TransUnion. If TransUnion you had limited the class at the front end to only people whose credit reports had been disseminated to third parties, you would have defined the class as the universe of people who were injured under this court's ruling. Then you probably could have just looked at TransUnion's records to figure out who was in or out. That is the polar opposite of a class like the one before you today. Thank you, Counsel. Mr. Yoshi? Mr. Chief Justice, and may it please the Court, this Court has frequently said that Rule 23 requires all class members to share the same injury. We think that includes an Article 3 injury. So if there are members of a class that aren't even injured, they can't share the same injury with the other class members. Respondents accuse us of somehow creating a special rule for Article 3 injury. I guess I view it a little bit differently. I view Respondent's Rule as creating a special rule for Article 3 injury because they would pluck Article 3 injury out of the certification context and either authorize or require, I'm not quite sure, district courts to delay and defer consideration of Article 3, and only Article 3, until the end of the case, after the merits, after the remedial stage, when it is time to dole out the actual relief. Rule 23 doesn't support that kind of rule. I don't think it's supported in practice as illustrated by cases in which the defense wins. I've never heard of a court certifying a class, ruling for the defense, and then figuring out if the plaintiffs have Article 3 standing. And I think it's inconsistent with the view of Rule 23 as a purely procedural aggregation device. I think my light went off. So what would you do at the early stages of the litigation, say, post-certification, and you find uninjured parties in the class? Yeah, so our view is surprisingly maybe not that far from what Respondent's are saying. I think the way we would approach it is what Rule 23 requires at certification is that the class be defined in such a way that on its face it includes only injured members. And at that stage of the litigation, you might not have much information about them, but then as the litigation proceeds, as AmChem recognized, courts have a duty to continually reevaluate the class. And if it comes to light that maybe there's a group of absent class members who aren't injured or don't share the same injury or really any other issue that might go to Rule 23, the court should reevaluate. Do I need to redefine this class to carve out those plaintiffs that I now know are uninjured? And the question then is going to be, can I do it in a way that doesn't require a lot of individualized analysis? And this is why I say I think we're not that far away from Respondent's here, and I think Petitioner agrees with us, that if there's some class-wide way or easily administrable way or mechanical way of identifying them, then that's what the court should do, and you can keep on going if there's not. If you're going to need, you know, 100,000 individual mini-trials... Why would that be the case, though? Because you have uninjured people in the party that you've now found. Why isn't that an Article 3 problem if it's an Article 3 problem up front at certification? So we are not making the Article 3 argument. We are saying Rule 23 is what requires commonality, predominance... So you don't think Article 3 requires injury? We have not taken a position on Petitioner's Article 3 argument. We're saying Rule 23 requires support. Okay, and if that's the case, let's work with your view, which is different than Petitioner's view, and I hear you're not endorsing it. We haven't taken a position on it. Yeah, okay. What do we do with historical practice where it was very common to treat in representative actions unnamed parties as not parties for purposes of the proceeding until and unless relief was given to them, and then you go through the injury analysis? I mean, your Justice Story's commentaries, for example. I guess I view the history a little bit differently. I think the historical examples, and, you know, we go through some of that in our own brief. I think in every one of those cases, it was obvious that everyone shared an Article 3 injury. Indeed, the representative action stemmed from the harshness of the rule in equity that all necessary parties had to be joined to a case. If you're a necessary party, you definitely have suffered an injury, and the representative action was meant to say it might be difficult to get all of those injured people, those necessary parties, joined, and so here's an exception we can create. So I read the history a little bit differently. It might be relevant to, say, a 23B1 class. That's sort of the forerunner, but I think 23B3 really is a 1966 innovation, and I think the further it strays from those roots, the more we ought to be careful about. Where do you find in Rule 23 the rule that the class must be limited to injured persons? We derive it from the Court's repeated statements and... How about Rule 23? Oh, from commonality, from typicality, adequacy, predominance. We think all of those tell us, as this Court has recognized, that all class members should share the same injury. I don't see why that would exclude the Article 3 injury at the core of the claim. But it's not an Article 3 injury. You say it's not an Article 3 requirement. It's a Rule 23 requirement. We're saying Rule 23 requires all class members to share the same injury, including, therefore, the same Article 3 injury. So it is an Article 3 argument, then. I'm just... I'm really confused now. I'm trying to help you out, so let me try and explain. We believe that under Rule 23, it requires that a class cannot be certified unless all class members share the same injury, including an Article 3 injury. So an Article 3 injury is required. It's a backdoor way of getting to Petitioner's position, I think. I think our approaches land at the same spot, but what I'm saying is that... So you think it's not required by Article 3, but Rule 23 requires Article 3 injury for all class members? Yes. I am saying Rule 23 requires it, whether Article 3... Where in the rule is that? I don't see Article 3 mentioned in Rule 23. No, but to say a class satisfies commonality and predominance is to say it has the same injury. That's this court's word. It's to say that overall, looking at the whole thing, it's manageable. There are at least some common questions. These name plaintiffs are generally typical. And common issues predominate. Every single person must have an Article 3. I don't get that out of the rule. The rule requires commonality and predominance. This court has interpreted those terms in Rule 23 to require all class members to share the same injury. That's why in Falcone the applicants claiming discrimination couldn't be certified in the same class with those claiming a denial of promotion for the same discrimination. That's why in AMCAM those exposed to the asbestos products who were ill couldn't be certified in the same class with those... That's the predominance issues and commonality issues, for sure. Yeah, and all we're saying is that Article 3 and Article 3 injury is the same kind of thing. If there are members of the class that don't even have an injury, how can they share the same injury with other members of the class who do? How does that satisfy commonality and predominance? That is our view of what Rule 23 requires. In other words, we're saying there's nothing special about Article 3 injuries. It should be treated just like any other element of class certification. That's our only... But Mr. Josie, when we look at commonality and predominance, and the treatises say this, they don't pick out a particular issue and say you have to have that. You don't have to have commonality with respect to every issue. So Justice Gorsuch's question is, if you don't have commonality with respect to the injury issue, what difference does it make? Why is that fatal to the class? There's no rule that says that particular issue you have to have commonality with respect to. I just want to supplement that. If you mostly have commonality with respect to the injury issue, but not with respect to every single person, what does that have to do with commonality and predominance? So let me take those in turn. Sure. So, I think this is a good question. experts' cases in Walmart, in Halliburton and Amgen, and in Comcast illustrate that there are some items on which, if there's variation across the class, they're so fundamental to the case that you really just aren't going to be able to certify the class. In Walmart, it was injury and causation. In Amgen and Halliburton, it's reliance on the securities claim. In Comcast, it was damages. And so the same argument could have been made in Comcast. We all had the same antitrust theory of injury, but because the damages were going to vary, that class couldn't be certified. And I think we're just saying the same thing. If you have a class in which Article III injury is not present for some, but is present for the others, that's just not going to meet the commonality standard. Now, Justice Kagan, you said, well, what if it's only a few? And I think my answer is the same as Mr. Francisco's, and frankly, the same that respondents give in their brief, or as I read it, which is, if there is a class-wide, manageable, mechanical way to separate them, as in, for example, TransUnion, there would have been, or as in Justice Jackson's Verizon hypothetical, there would be a manageable, class-wide way to do it. We think that's fine. Rule 23 doesn't preclude that certification, but what we are saying is that Rule 23 needs to be followed at certification and then throughout the litigation. As the litigation proceeds, if there's more infotainment... Well, hold on. I thought you said that commonality means... I'd understood it as one issue has to be common, and that that has to be predominant. That has to be predominant. That's the way I understood it. Okay, fine. No. You're telling me that Article III and Article III alone must be satisfied by everyone at the outset, I thought. What I'm saying is Article III injury is no different from any other requirement for class certification that should be common to the class, like injury, causation, reliance, damages, all of these must be common. And I take the point that... They don't all have to be common. There has to be a common question that predominates over others. And now, a special rule that Article III must exist for all class members? I am not asking for a special rule any more than Comcast had a special rule for damages, or Walmart had a special rule for causation, or Halliburton and Amgen had a special rule for reliance. I'm just trying to say that Article III injury is of that sort important enough that it's just unlikely you're ever going to be able to certify a class. Thank you, Counsel. I'm not sure I've been following the bouncing ball. Does Article III in this context require an injury? Now, I don't know if you're saying that not at the outset, but at the back end before any relief is granted, or what? Article III of its own force, of course, requires every class member to have standing before he or she can collect a damages award. That's transunion. I'm happy to stop there. Yeah, well, so I take Petitioner's point to be that... Or Petitioner's Article III argument, as I understand it, is that Article III of its own force also requires each class member to demonstrate standing at certification. And we're not taking a view on that. What we're saying is that Rule 23's commonality and predominance requirements requires that same thing, and so there's no need to decide whether Article III of its own force would require it if, say, Rule 23... So Article III is in the case. You just like to run it through the certification process? That's right. Okay. Justice Thomas, anything to add? Justice Alito? Well, just to clarify this last point, are you saying that Rule 23 requires something that just happens to correspond with what Article III requires? That I would understand. Or are you saying that Rule 23 requires compliance with Article III, which then doesn't seem to me to be any different from Petitioner's argument? Yeah, we're saying the first thing. And all we're saying is that as an empirical matter, in practice, an Article III injury is just so fundamental to the claim that just like in Walmart or Halliburton and Comcast, it's the kind of thing that if it's not common, if it's individualized, then that's probably going to predominate in such a class. Okay. So why do you want to... Why do you approach this issue in that way? Is this just sort of abstract respect for constitutional avoidance or does the government think that there's some... that there are different consequences from taking your approach and the Petitioner's approach? No, it really is from constitutional avoidance and ruling on this case narrowly as the case presents itself. We took the Court when it reframed the question presented to limit it to 23b-3 as a signal that maybe it wanted us to talk about Rule 23b-3 and that's what we think you should do here. Thank you. Justice Sotomayor? Transunion. I can go back. I'm relying on old memory. But I think the class was defined as anyone who had false statements in their credit reports. It wasn't through until the litigation came forward that we found out that some people's false information was not disseminated. And we basically said you can't give out the damages to the people who weren't injured because there was no dissemination. But that wasn't known until the end. I think what Mr. Francisco is now saying and I'm not sure you are or aren't, that now we have to have that fight at the class certification stage that we have to define a class in a way that says only people who receive the report instead of the way it was defined. Do you agree with that? Now that we know. No, but to use with that, we don't know at the beginning. But he's going to put an affidavit in that says some weren't disseminated so this class shouldn't be certified. If we don't know, then no. I mean we're not asking for rule 23 to be applied in a senseless way. We think it should be applied sensibly, reasonably, with reasonable inferences. But now that we know, now that we have TransUnion on the books, if there's a future claim under FCRA for OFAC warnings on credit reports, yeah. I think a court there should say well I know in TransUnion this class of plaintiffs wouldn't have standing so I'm going to certify a class only of plaintiffs whose credit reports were disseminated. That would be the responsible thing to do now that we have TransUnion on the books. But our view is that in a case like TransUnion or in a case like Tyson Foods, which I would love to talk about We go back to is there an admissible way to identify the injury? Exactly. And that's what the 23F inquiry is. 23B3. Exactly. Do you think, Mr. Joshi, that this is the way we've handled class actions over the years? If you look back the last 70 years of class action, whatever Rule 23 is, it strikes me that if you look at all the classes that have been certified by that point, you're always going to be able to find people for idiosyncratic reasons who don't share the same injury, who don't have standing, and that that's never been seen as kind of the end-all and be-all, the whole, like, okay, we have to explode everything. So it seems very inconsistent to me with the way class actions have been practiced for many decades. Yeah, so I disagree. I have read every single one of this Court's class action cases, you might imagine, in preparation for this case, and the one theme I see consistently is that where there's a difference in injuries or the type of relief or even the type of remedy that, you know, the defendant is requested to make, this Court has said that those really can't be in the same class together. And it just strikes me that Article 3... The rule that you're suggesting is a rule that says to district courts, you have to do this right up front, you have to figure out whether everybody has the exact same injury. If, like, there are a few people who have a different kind of injury, that's verboten. I mean, that is something that I don't think district courts have ever thought that they needed to do. Rather, what they thought is by the time we get around to issuing remedial orders and issuing damages, we better make sure that we're not handing out money to people who aren't injured. So that, I think, everybody has understood as their obligation. But not this. Not, like, we have to do all the work the moment the case comes to the door to figure out exactly who is injured and how. I guess I have a few responses to that. One is what I just said to Justice Sotomayor, which is that we think Rule 23b3 should be applied in a sensible, reasonable manner. So, for example, if there's say, an antitrust class and the allegation is price fixing and the class is defined as all purchasers of the product during the period of price fixing where there were super competitive prices, we think that would be a valid class definition, even if there's some idiosyncratic person that likes paying higher prices for whatever reason. Or if it's a product liability, you know, a defective product that injured people, you know, all purchasers of the product who suffered the injury would be a valid class in our view, even if there's somebody who, because of the injury, you know, missed a test and then that he didn't study for but then did the makeup test later on, got a better grade, got a Supreme Court clerkship at the end of it, and therefore wasn't injured. You know, those sorts of idiosyncratic things, we agree. That's not what Rule 23 requires. Okay, so this is really not an Article 3 rule, because if it were really an Article 3 rule, you couldn't agree on all those things. That's right. We think this is a Rule 23 issue. It's just that when we read the question presented as reformulated, the court was very careful to say an Article 3 injury, and we read that as saying, how is that different from the kinds of injuries in Falcone you said couldn't be Yeah, I mean, but I'm taking from your thing, you know, you went back and forth with Justice Gorsuch about were you endorsing, were you not endorsing, do you have a position? In fact, you do have a position on Mr. Francisco's hard Article 3 argument, because you couldn't have said that those classes should go forward if you accepted Mr. Francisco's argument. We're saying those classes could go forward under Rule 23. We are not taking a position on whether Article 3 of independent force would prevent that, would preclude those sorts of classes. We don't think it's... Come on. Okay. That's our position. Justice Gorsuch? Yeah, there are a lot of price-fixing cases where the victim can pass through the overcharge and suffers no injury, but you let that go forward. That's a substantive rule of antitrust law, I believe. It's a fact in the world, too. And they're not injured, and you'd let that class go forward. As I said, my understanding is that the substantive rule of antitrust law that only direct purchasers can bring claims. Well, in some places yes, in some places no, and after Apple, I don't know, but you would allow that class to go forward, no Article 3 problem. Under Rule 23, there's no problem. Okay. All right. I just want to make sure I understood it. You've heard some of the discussion about the procedural problems in this case. The government didn't talk about them in its brief. I wanted to give you an opportunity to give us your thoughts on whether we have the problem before us, given that the May order talks about only injured persons. Yeah, so we didn't talk about our brief because our brief was filed before the red brief was filed, so we didn't know this issue was going to be raised. It wasn't raised in the brief in opposition. As an amicus, we're poorly situated to take a strong view of matters, but that said, I think petitioners have the better of the argument, and what I would rely on are two things. One, under cases like Sarko v. Kadish, we know that the Court of Appeals issued an adverse judgment to petitioner. It's jurisdictionally properly before this court. This court has jurisdiction over the case, so I don't think it's a matter of jurisdiction. So then there's the question, well, which order are you really looking at here? And I guess I would place greater emphasis on something that Mr. Francisco mentioned a couple of times in his discussion, but for me it's very important, which is on page 63A of the appendix. This is the August order. The August order does not purport to enter a new class or certify a new class and get rid of the old one. What it says is page 24, lines 13 to 23 of the earlier order is replaced with the following. And so I view it as sort of nunk-pro-tunk, modifying the earlier order for which there was a notice of appeal, and I know there's been a lot of... I appreciate that. What do we do about the fact that the only order that the Ninth Circuit was reviewing was the May 12th order, and it specifically said we can't hear. We have no authority, no jurisdiction over the August order that you're asking us to rely on. You didn't address that squirrelly complication. Yeah, again, you know, there's an August order, but one of the terms of the August order was to nunk-pro-tunk... I understand, but the Court of Appeals didn't understand itself to have jurisdiction over that order, and we're only reviewing the Court of Appeals resolution of its view on the May 12th order. That's right, and that's why I would fall back on the fact that you have jurisdiction to review the Ninth Circuit's decision. We think it contains an error of law. You could correct that error of law, and, you know, Justice Barrett asked, what's left on remand? I think if you corrected that error of law and sent it back, even if the Ninth Circuit adheres to its view that it had only the May class definition in front of it, and even if the Ninth Circuit then holds that the May definition doesn't run afoul of the rule that, by hypothesis, he would adopt in this case, then, and even if, as Mr. Francisco suggested, he's out of time to appeal the August order, I would imagine that on remand in the District Court, with that binding precedent, Petitioner could move for decertification or... I accept all of that, but that really does sound sort of like an advisory opinion to me, because the only binding force we would have is to say that the May 12th order was fine because it addressed only injured people. Period. I mean, that's our judgment. This Court frequently corrects errors of law in what Court of Appeals say without analyzing whether the prevailing party below could nevertheless still prevail under the new rule. It does that all the time, and I think you could take that case here. Take that approach here. When you said Rule 23 is narrower than Article 3, I just want to make sure I understand how you think it's narrower. It's not narrower in the result that would be reached in particular cases, at least as I understand your position. It's narrower, I gather, in the sense that theoretically Rule 23 could be changed, and at that point, we would have to confront, in your view, the Article 3 issue. Is that what you mean by narrower? That's exactly what I mean, yeah. Okay. Thank you. Justice Barrett? Justice Jackson? I guess I'm still struggling with why isn't the sensible and reasonable manner of applying Rule 23 the way that, as Justice Kagan says, we always do, that we look at commonality or that district courts look at commonality and predominance sort of in the aggregate with all of the issues on the table. There aren't directed to one or another to say that if this particular issue does not have commonality, you can't certify the class. And that seems to be what you are saying. I appreciate that there are certain other cases where the court has picked out various issues and said either you've gotten it wrong or right on commonality, and that you would like for this to be one of them. But it's unclear to me that the rule is such that it requires that this particular issue, there has to be commonality with respect to. I think the one thing I read the cases differently. I think there is a strong through line of this Court's class action cases Falcone, Amchem, Wal-Mart, Lewis against Casey. So many cases, East Texas Motor Freight, in which the Court has said that different injuries cannot be certified in the same class. Let me ask you a question. Do those cases talk about those injuries in the context of the harm being an element of the claim? If I go back and look at them, are those cases ones in which the harms that we're talking about are just in the damages realm? There are many claims in which harm is actually an element of liability, and I totally understand in those worlds you're thinking about, can this be proven by common proof, or do we have a bunch of individual actions here? But it seems to me that when we're talking about damages apart from liability, it's very hard to see a world in which individual proof with respect to damages can overwhelm from the standpoint of predominance the kind of consideration of whether or not you should have a Rule 23 certification. So there are cases of both types, and Comcast is the perfect example of the case you just said. In Comcast, the plaintiffs had an antitrust theory that was common to the class, that was common to the defendants, that would have established the injury causation, etc. But it was the variation in damages that precluded certification of that class, because it's just so fundamental to the kind of claim that was being brought, that it was just going to overwhelm even the common antitrust liability theory, and we're saying Article 3 injury is just as fundamental, and Rule 23 would preclude certification in those circumstances. Thank you, Counsel. Mr. Drifter? Mr. Chief Justice, and may it please the Court, as this Court held in TransUnion, and as the Chief Justice recognized in Tyson Foods, Article 3 doesn't give federal courts the power to give relief to any uninjured plaintiff, class action or not. So if the Court finds its way to reaching the question presented and writes an advisory opinion, and that's what we think it would be, the advisory opinion should hold that at the class certification stage, the proper inquiry is whether there will be an administratively feasible mechanism to weed out the uninjured. Consistent with centuries of historical practice from the Chancery Courts at the time of the Constitution's ratification to now, it is the representative who is actually before the Court, not the absentees, who must prove the existence of an Article 3 case or controversy at the outset. But if we are here to police the jurisdiction of the federal courts under Article 3, we should probably start with this case in this Court. Wab Court now concedes that any appeal of the August order on which Wab Court's arguments have exclusively relied was not actually in the case in the Court of Appeals, and therefore isn't within this Court's certiorari jurisdiction. In its reply, Wab Court has shifted gears and attempted to reorient its challenge to the May order that is concededly no longer in effect and that is not harming Wab Court. But any appeal of that superseded order is moot. The traditional exceptions to mootness do not apply and this Court should reject the invitation to craft a new mootness exception on the fly. It would make little sense for this Court to reach broad pronouncements on Article 3's limits in a case that itself presents such serious jurisdictional barriers to reaching the question presented. I welcome the Court's comments. Did you raise that in your brief in opposition? I'll acknowledge that the brief in opposition did not raise this jurisdictional problem. It did raise the fundamental problem that in the respondent's view the question presented is not presented by this case and other preservation issues. But as the amicus brief of the federal jurisdiction professors indicates, there were a number of ways in which the petition obscured this jurisdictional problem. It became apparent when the blue brief was filed that the arguments rested entirely on this outdated order and class definition and we pointed it out in the red brief. In hindsight, with the 2020 view of hindsight, I wish all of these issues had been fleshed out at the cert stage and perhaps we wouldn't be here. But I do think even when wisdom comes late, we should acknowledge it and this Court's rules are clear that the failure to raise jurisdictional objections at the certiorari stage doesn't deem them forfeited. What's your view of what would happen on remand? I asked Mr. Francisco what would happen given that the 23F time has expired. You heard the question. What's your view of what would happen and whether that August order remains something that he could appeal or not? Right. And so we don't think there should be a remand if you agree with us that the case you could dismiss as improvidently granted or perhaps it would be vacant. Even if you did that, then it goes back to what happens after that. Yeah, yeah. So I think they have available to them the ability to move to decertify the class. They have the ability to move to dismiss for lack of standing the named plaintiff. So they're not without rights. And Rule 23F is a discretionary appeal mechanism and there was some discussion earlier about what was ascribed to the Ninth Circuit, this material change doctrine. It's actually not just the Ninth Circuit. I know you don't have briefing on this, but all of the circuits have had to confront this question and I think their approach is similar which is it doesn't assign dispositive significance to what the district court said, whether the district court characterizes there being a material change. The Court of Appeals has the discretion under Rule 23F to have that gate keeping role and to decide whether there's an appeal. So it's true that they failed to take an appeal under 23F from that August order, but it would have been a mistake to ascribe any significance for the district court's determination on this and actually in our opposition to the Rule 23F I just want to make sure I point this out, Justice Alito, that in our opposition to the Rule 23F petition, we did point out that there was a fight in the district court about the class definition and that could actually render the Rule 23F petition moot. So we put the defendant on notice of this. They had the ability to file a second petition and they didn't and they still have the ability in the district court to, because class certification is always a constant moving target, they have the ability to seek relief even now in the district court and then perhaps if they don't like that they can appeal and maybe they would come back here. Sorry, Justice Alito. Do you think this material modification rule is required by statute? By Rule 23F? I think Rule 23F In other words, if a different court of appeals said we don't want any part of this rule, we think it's a silly rule, we're going to adopt a different rule. Would that be wrong? I think that Would it be contrary to some statute? I think the best way to understand what the court is saying is that they are interpreting Rule 23F and Rule 23F limits the interlocutory appeal jurisdiction to an order granting or an order denying class certification. And so the courts are trying to figure out do we have a new order granting or denying certification. And in this case I think it's quite clear and actually I disagree with Mr. Joshi's characterization. I think it's quite clear that the May order and the August order are different orders with respect to class certification. In fact, the August order, it's true that it says it's modifying the previous order. That was the June order which the Ninth Circuit also said was outside of its jurisdiction. Suppose the district court issues an order certifying a class. There is an appeal. And after that, the district court makes some change in the order. Like the district court says, look, this is not material. This is immaterial, absolutely immaterial. What is the party who's taken the earlier appeal supposed to do? The party would, I would think, if your position is correct, the party has to say, well, you know, I don't want to bet everything on this. Even though the district court has said it's immaterial, I'm going to have to file a new notice of appeal always. Right. So I think that's exactly what this doctrine is intended to address. And I think, you know, if there was a typographical error, for example, in a class certification order, I think nobody would reasonably insist that there needs to be a second 23-F petition. And I think Judge Posner has a decision on this in the Apple, Illinois case. I think he explains that that's part of the reasoning for this. But I just want to point out, we're here, you know, in this case talking about 23-F appealability jurisprudence in a case where my friends didn't challenge the jurisdictional holding of the Court of Appeals. If they had taken, if they had filed in their cert petition, if they had told you about this problem, and they said, look, there's a May order and an August order, and our beef is really with the August order, and maybe we think the jurisdictional holding of the Court of Appeals was wrong, they could have challenged that, but they didn't do that here. I think the reason we're talking about it is, as I understand the reply brief and the letters that were exchanged, I think the suggestion, LabCorp's suggestion is that the case is not moot because you should import — this is how I understand the argument — that you should import this material change doctrine into the mootness exception and craft some kind of exception for mootness on that basis. And they cite in their reply brief the Jacksonville case. That case is a case about an exception for mootness, but it is about the voluntary cessation and capable of repetition exception, which is a recognized exception. I don't want to belabor this, but I will ask one more question on it. If the Ninth Circuit's rule is not required by statute, then why is there a jurisdictional problem? Oh, I think that the Court of Appeals has to have some way of determining what is within their jurisdiction and what is not. And remember, it's discretionary. So they have certiorari-like — Well, that's a different question, but surely the Court of Appeals can't by means of a decision say we are defining our jurisdiction in a particular way? Well, I think they exercise their discretion with respect to an order granting or denying class certification. They exercise that discretion with respect to the order that they were asked to review that was attached to the petition. And then the question is should some kind of exception be made because there was a subsequent order that wasn't appealed? Thank you. I think you said earlier, and I might have misunderstood, but I think your position was that the petitioners should have disregarded the district court's characterization of its own order. Is that correct? Well, yeah. I mean, that's not quite how I'd put it because I think — I think that's not what you said, but I think that's what you mean. And I doubt many lawyers — you know, your clients — I doubt, you know, anyone really wants to live under that rule, that a lawyer should be disregarding how the district court characterizes its own orders for purposes of these timing rules. That strikes me as asking for a lot of chaos. I think as an earlier colloquy with Justice Kagan brought out, a lot of people said a lot of different things about the differences between these orders, and perhaps it'd be worth pointing to what the defendants said when there was a fight about this order, and this is in the district court at document 110 on the first page of their brief about the refinement. They said this is no refinement at all. This proposed definition, the August definition, is broader than the existing one and clearly includes those who have no injuries, so they understood it to be a very big change. The question was about how the district court characterized it. Right. I don't think — I don't read that — At least at times, the district court characterized it in a way that you say, oh, well, a reasonable lawyer would have just ignored that, and that just strikes me as contrary to how lawyers practice law and just say, oh, well, the district court is clearly wrong in how it's characterizing its own order, so we should just ignore that and file this and that. Well, that one footnote wasn't the only thing that was said, and I think if I were advising a client in this circumstance, I would say, look, if what we really want to challenge is the August order, we better make darn sure that we challenge the August order, and we shouldn't rely on the fact that the order that we regard as — Would it be held to that standard, always? Yeah, I mean, I don't think this is a close question under the lower courts' material change doctrine cases, and I think another thing that Judge Posner said in that case that I mentioned is that what the inquiry turns on is what it is that the party seeking the 23F appeal is actually seeking to challenge. It was pretty clear that there was a fight over these definitions and that LabCorp regarded this as a big change. Recall that they persuaded this court to grant certiorari on the idea that you've got unscrupulous plaintiff's lawyers that are stuffing classes full of uninjured people, right? But they regarded that first definition, the problem with it is that it was actually too tethered to the plaintiff's injury, that it was fail-safe because it only had uninjured people, and then they regarded the second definition as broader and wanted to challenge that definition. So under those circumstances, I do not think it would be reasonable for somebody to rely on the idea that the original 23F petition didn't extend. But we're now here talking about the appealability of that order, and that jurisdictional holding wasn't challenged. And I actually read the reply and the letters that were exchanged as acknowledging that LabCorp hasn't preserved and isn't seeking to contest that jurisdictional holding of the Court of Appeals. So now what you're left with is an appeal from an order that has been superseded. And I alluded to the general rule in this Court's cases. And this comes up when you have, for example, a preliminary injunction that has been outstripped, and then you had an appeal from the preliminary injunction. This Court has said those appeals are moot. Or if you have, for example, an appeal with respect to a complaint, the complaint has been amended, the interlocutory appeal is rendered moot. You might have a case, a redistricting case where there's a debate about a map, and then the map has been changed. That appeal would be rendered moot. And so that's the general rule. And that's why I said earlier that what I regard LabCorp as asking you to do is to craft an exception from that general rule on mootness for the circumstance. And I think I regard them as relying on this material change doctrine from the lower courts as supplying a standard for that mootness exception. Mr. Francisco can correct me if I'm wrong, but that's how we read their reply brief and the letter. And I think, you know, in our view, that would be an ill-advised thing to do. You don't have briefing on that. And as the examples that I recited, I can tell you this is not an unimportant question. It is something that is recurring. And even in class action practice, I think this is a recurring issue about how the courts of appeals police the boundaries of their jurisdiction as class actions are continuing to move through the district courts. And it's important that jurisdictional rules be clear, to be sure. It is also important that the court of appeals be able to use their limited resources to exercise their discretion to decide live controversies with respect to actual in effect class certification orders, rather than have appeals that are backwards looking and are about a target that is already moved. In their opening brief, petitioner said, quote, the definition for the damages class, the only class before this court is as follows. It gave the August definition, not the May definition. Mr. Francisco relies on, and I think it was the question that Justice Alito was referring to, that they viewed the district court class definition as not meaningfully different between the May and August. But it was meaningfully different because of your change, correct? Your change was in response to their claim that you had a failed class definition. The reason they're pointing to statements by the district court or by the plaintiffs is if you actually look at the lab court has been fairly consistent that they regard this as a big change. And they regarded the original definition, as I said earlier, as too tethered to injury because it was defined in terms of who was denied a full and equal services. This is the hard question. It may be unfair and you can tell me you want to think about it. But they claim that you do not have an administrable way of identifying the injured and uninjured. So whether it's under Article 3 or it's under Rule 23 according to the SG that you can't prove that. What's your point on that? Well, so, you know, I'd be happy to talk about it in the abstract and I can talk about all the lower courts have said in cases where this question has actually been presented. The oddity of this case is that issue was never presented in the district court. And so the district court didn't have any battle over this and didn't certify a class that was premised on the idea that there was a contest over whether there were uninjured people. And actually our position has been all along that everyone in this class is injured and that's what the lower courts I think recognized. And the reason why is similar to other cases where there's a discrimination alleged. This court has always said discrimination itself is an Article 3 injury. And so one analogous case is a case where you have people that are challenging the affirmative action policy of a university. Well, Mr. Gupta, that's the question that we didn't get to in Ashton, right? There are arguments that racial discrimination and other kinds of discrimination are different. So I do think in fairness that that's an open question. Well, I mean, you don't have a case directly in point and I agree that was teed up in Ashton. I think the case is very, very different from Ashton because these are not people who are in Maine talking about something in Hawaii. I'm not saying it's the same and maybe you might win. All I'm saying is I don't think it's as subtle as you're presenting it. Sure. And it wasn't presented in the lower courts and so that's why I'm in the position of just kind of making this argument on the fly. But I'll say if you set aside, and you're right, Justice Barrett, if you set aside this question of whether disability discrimination maps on to this court's precedence on discrimination, and we think it should, I think if you do that, this is a case where all of the people are injured for the same reason as in a case like Gratz v. Bollinger, the affirmative action case where what the court said there was you were confronted with this barrier on the basis of the protected characteristic, the fact that you didn't reach the thing. I didn't mean to lead you down this road because that's not before us, right? We didn't take that. It is not. And so the whole point is that even if we assume that you're right and that the class as you defined it does include only people who are injured, that doesn't take away Mr. Francisco's argument that there would still have to be some sort of process certification to identify who was injured or not, even if it was just who wanted to go to lab court. Well, I think Justice Barrett, I mean, this points up the strangeness of this vehicle because this is a question that arises with some frequency in the lower courts, but where it arises, there is an understanding that there's a real question about whether there are uninjured people and how they will be weeded out. And it principally arises in cases, it happens a lot in antitrust cases and other kinds of cases where the plaintiff's method of proof relies on economic model about a counterfactual world. And so, you know, in a price fixing case, for example, there's going to be a question, did everyone pay the super competitive price? And it might not be possible ex-ante to determine who the people are. But what you're saying, I mean, because I don't want to take up your time and you can divert it, just what you're saying is that it would be impossible if we agree with you on kind of the Atchison-esque point, you're saying it would be impossible for everyone in the class not to have standing as you describe it, so that this would just kind of be irrelevant. Yeah, I mean, it's another way in which writing an opinion here would be an advisory opinion because you'd be... Well, we'd have to decide that we agreed with you on the question that we don't want to answer, which is... Well, you'd have to reach an antecedent question that really wasn't presented below. We deliberately excluded it from the question, so... Right. I mean, I think you do that all the time. You grant cases where there's an assumption built into the question presented, but I think it does matter whether that assumption is true or has been established in the lower courts. I don't want to fight on this, you know, because it's outside the QP, but I would just to continue what I was saying, I think where the question arises, it tends to be where there's a battle of the experts, as Justice Gorsuch was talking about earlier in cases like antitrust cases, or in a case like Tyson Foods. Tyson Foods was a case that came here where you had a complex question about how to weed out the uninjured people, and because the defendant hadn't kept records, the plaintiffs had to rely on expert testimony, and there were 212 people at that pork processing plant in Storm Lake, Iowa, who it turned out were not injured. It would have been really easy to have a trial and weed them out if there had been a special interrogatory form. Those were people who were uninjured simply because they didn't work over 40 hours, and so they weren't deprived of overtime. But if you accept the submission on the other side here today, I think what should have happened, in their view, is you would have redefined the class, and then you would have had a fail-safe problem, perhaps, which is you would have defined the class to be only those people who worked 40 hours and were not paid overtime. That's not really in the defendant's interest. The defendant, as Justice Gorsuch was describing earlier, the defendant wants a class definition that's ultimately going to provide global peace when the case is resolved. What I said in the short concurrence, which Justice Alito and Justice Alito alone joined, that the answer, when you get to the point and realize that these people had not worked 40 hours, is not that you then go back and carve them out. What I suggested is that that would be a good reason not to certify the class in the first place. I think the problem in Tyson Foods, though, was you recall the conundrum that the parties faced there was because they hadn't done a trial where there was a special interrogatory. It was a wall-one proceeding, so you had this weird aggregate judgment, and then the question was is there some way to reverse-engineer what happened with that judgment to weed out the uninjured? But if you had to redo the experiment of Tyson Foods, it's a pretty standard kind of case, except that the defendants didn't have those records, and it would have been easily possible to try that case in a way where the case would have weeded out those uninjured people, and then of course would comport with the defendants seventh amendment rights. Everybody knows that the elephant in the room that once you get to trial, the possibility of facing the damages that are at issue in many of these cases is enough to prevent the defendants, as a practical matter, from going to trial. Well, I mean, I acknowledge that many class-action cases don't go to trial. I'm not sure that's true of litigation in general, but I think the suggestion on the other side is that the interim effect of certification is magnified because of the incremental difference in the class definition that includes some uninjured people, and the story is, right, that's the story on the other side. This case itself belies that story, right? The problem here was the plaintiffs, in their view, had defined the class in a way that was too tethered to injury, and they wanted to change that class definition to expand it, and at the end stage of litigation, it is the defendants who want the broadest possible definition. I'm sure there are situations they should do this because it's actually going to be good for them, but on the other hand, it's reasonable to suspect that that's not always going to be the case, and that they may be the best judge of whether it's good for them or not. I mean, look, I think in litigation where there's high stakes, the parties are going to behave strategically in a way that maximizes their interests, and as I'm suggesting, that interest changes over time, but I just resist the suggestion that what's really happening in the real world is that there are some incremental marginal advantage that plaintiffs are seeking to get by expanding the definition to include the uninjured because that just creates—the goal is to try to eliminate those manageability problems to the extent possible, and in the settlement calculus, those uninjured people, we know after transunion, they're not going to recover, and so I think the policy argument—I'm not sure that this is the right body to be considering those policy arguments, but I also think the economic logic just doesn't hold up. Do you think the amicus briefs are wrong, then? They're just not understanding their own interests? The amicus briefs on the other side—I guess I'm picking up on the Chief Justice's question. I think they know their own interests. I'm not saying they're right. I'm just saying you're saying their interests are just misguided entirely. Well, I actually think some of the amicus briefs on the other side take a more measured position that is really more consistent with the consensus view in the lower courts on Rule 23. The parties' positions have coalesced quite a bit on the Rule 23 question, and I know you were persuaded to take this case on the idea that there's some circuit split, but if you actually look at the circuit decisions that are deciding this question on Rule 23, I think the divergences are largely explained just by the differences in the record and the economic models, which are complex. But they're all saying really the same— I think Judge Katsas and Judge Ikuda and Judge Dyke in the First Circuit—they're all saying the same thing, which is this has to be administratively feasible, and we have to figure out whether it's going to be possible— Do you agree with Judge Katsas's opinion? Yeah. I mean, I think I might— Perhaps if I were sitting with Judge Katsas on that case, I might have come out differently on those facts, but I think the legal framework for these cases, they're just not very different, and I think everyone recognizes, especially after TransUnion, that the job is to weed out the uninjured, and it's just a question of whether on those records, whether it's going to be manageable to do so. I think the Article III argument on the other side here is much more ambitious and would really be a departure from the way things work. If I could just— Mr. Francisco's understanding of this case is you have sort of two groups of people, the ones who wanted to use the kiosks, who tried to use the kiosks, who couldn't use the kiosks, and the ones who wanted no part of the kiosks. And, of course, that's very different from your understanding, which is discrimination is discrimination. But just take for a moment— and this is a question that we're not going to decide one way or the other in this case— if you take for a moment Mr. Francisco's understanding of who has, you know, what the weed and what the chaff is, is he right that you have no way of separating out those two groups of people? No. I think it would be a harder case than this one, but I think it's not infrequently the case that, you know, membership in a class turns on some attribute of a person that can be tested through a claims process. And you have an amicus brief from the claims administrator that explains how this happens. It happens in a lot of different contexts—products, liability—and there can be, you know, there's discussion of affidavits. There can be affidavits. That can be one way it can be done. It can be done based on an examination of records. The defendant often has records that will confirm membership in the class. So I reject the suggestion that that's impossible to do, but I think, you know, as this Court said in Dukes, like, that the predominance inquiry is very case-specific and it has to be a rigorous inquiry based on the record. And so what I appreciate, you're not going to answer, you know, the specific standing question in this case, but I would also just caution the Court because of the strangeness of this vehicle, where none of this was teed up in the courts below, not to paint with a broad brush and address situations that aren't before the Court, where actually managerial district judges a [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A98591. Public record. Not legal advice.